Parallel-in-Time Power System Simulation Using a Differential Transformation Based Adaptive Parareal Method

被引:2
|
作者
Liu, Yang [1 ]
Park, Byungkwon [2 ]
Sun, Kai [1 ]
Dimitrovski, Aleksandar [3 ]
Simunovic, Srdjan [4 ]
机构
[1] Univ Tennessee, Dept EECS, Knoxville, TN 37996 USA
[2] Soongsil Univ, Dept Elect Engn, Seoul 06978, South Korea
[3] Univ Cent Florida, Dept ECE, Orlando, FL 32816 USA
[4] Oak Ridge Natl Lab, Computat Sci & Engn Div, Oak Ridge, TN 37830 USA
关键词
Computational modeling; Mathematical models; Adaptation models; Power system stability; Convergence; Numerical models; Power system dynamics; Differential transformation; Parareal; parallel algorithms; power system simulation; power system stability; DYNAMIC SIMULATION;
D O I
10.1109/OAJPE.2022.3220112
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
For parallel-in-time simulation of large-scale power systems, this paper proposes a differential transformation based adaptive Parareal method for significantly improved convergence and time performance compared to a traditional Parareal method, which iterates a sequential, numerical coarse solution over extended time steps to connect parallel fine solutions within respective time steps. The new method employs the differential transformation to derive a semi-analytical coarse solution of power system differential-algebraic equations, by which the order and time step, as well as the window length with a multi-window solution strategy, can adaptively vary with the response of the system. Thus, the new method can reduce divergences and also speed up the overall simulation. Extensive tests on the IEEE 39-bus system and the Polish 2383-bus system have verified the performance of the proposed method.
引用
收藏
页码:61 / 72
页数:12
相关论文
共 50 条